The epidemics of delinquency, violence, school drop-out, and drug abuse have been studied in regard to antecedent risk behaviors and conditions. Specific early risk factors are now being targeted by prevention programs in randomized field trials. In recent years an important structure for prevention science has emerged based on progress in research on early risk and protective factors along developmental paths as early as first grade and possibly earlier. These have been designated targets for preventive intervention trials here in Baltimore and elsewhere (Durlak &Wells, 1997;Gersten, Seals &Kallgren, 1991;Kellam &Van Horn, 1997). This prevention research strategy is based on an integration of three scientific perspectives: The first is life course development with a focus on early risk and protective factors and developmental paths;the second is community epidemiology with a focus on defined populations in their environment, variation in developmental paths, and control of selection bias;and the third is experimental interventions precisely targeting early risk factors in developmental models (Mrazek &Haggerty, 1994;NIMH, 1993;Kellam &Rebok, 1992). This form of prevention science is multi-disciplinary, requiring the coming together of biological, psychiatric, psychological, social and cultural, economic, and biostatistical disciplines. The numbers of researchers is severely limited, and the need for training the next generation of prevention researchers is very high priority (Mrazek &Haggerty, 1994;NIMH, 1993). This application is for a five year continuation of the Johns Hopkins Prevention Research Training Program, that grew out of the Johns Hopkins Prevention Research Intervention Center and its first two generations of preventive trials. Candidates will continue to be recruited from disciplines including psychology, psychiatry, sociology, biology, biostatistics, and economics. Prevention research training will continue to be centered on developmental modeling, design of epidemiologically based randomized field trials, biostatistical methodology, and designing partnerships with community and institutions for randomized field trials. More intensive training will be done during this next five years on the neuropsychological and genetic underpinnings of behavior, the role of community factors in development and psychopathology and as moderators of intervention response, and economic analyses of preventive interventions.